Early-Signal Detector

method · first seen 14 Oct 2026

reinforcement learning from measured performance for code optimization

Training a code-generating model with a reward combining correctness and measured runtime speedup of generated kernels.

Mentions per week

0119 May11 Aug10 NovPapersCodeDiscussion

PapersCodeDiscussion

Earliest sources

Latest sources

Mentioned together with

GPU kernel optimizationTVM MetaSchedule